Subject(s)
Lipoma , Liposarcoma , Thymus Neoplasms , Humans , Liposarcoma/diagnostic imaging , Diagnosis, Differential , Lipoma/diagnostic imaging , Thymus Neoplasms/diagnostic imaging , Tomography, X-Ray Computed/methods , Male , Female , Middle Aged , Contrast Media , Magnetic Resonance Imaging/methodsABSTRACT
PURPOSE: To develop and validate a computer tool for automatic and simultaneous segmentation of five body tissues depicted on computed tomography (CT) scans: visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), intermuscular adipose tissue (IMAT), skeletal muscle (SM), and bone. METHODS: A cohort of 100 CT scans acquired on different subjects were collected from The Cancer Imaging Archive-50 whole-body positron emission tomography-CTs, 25 chest, and 25 abdominal. Five different body tissues (i.e., VAT, SAT, IMAT, SM, and bone) were manually annotated. A training-while-annotating strategy was used to improve the annotation efficiency. The 10-fold cross-validation method was used to develop and validate the performance of several convolutional neural networks (CNNs), including UNet, Recurrent Residual UNet (R2Unet), and UNet++. A grid-based three-dimensional patch sampling operation was used to train the CNN models. The CNN models were also trained and tested separately for each body tissue to see if they could achieve a better performance than segmenting them jointly. The paired sample t-test was used to statistically assess the performance differences among the involved CNN models RESULTS: When segmenting the five body tissues simultaneously, the Dice coefficients ranged from 0.826 to 0.840 for VAT, from 0.901 to 0.908 for SAT, from 0.574 to 0.611 for IMAT, from 0.874 to 0.889 for SM, and from 0.870 to 0.884 for bone, which were significantly higher than the Dice coefficients when segmenting the body tissues separately (p < 0.05), namely, from 0.744 to 0.819 for VAT, from 0.856 to 0.896 for SAT, from 0.433 to 0.590 for IMAT, from 0.838 to 0.871 for SM, and from 0.803 to 0.870 for bone. CONCLUSION: There were no significant differences among the CNN models in segmenting body tissues, but jointly segmenting body tissues achieved a better performance than segmenting them separately.
Subject(s)
Deep Learning , Humans , Tomography, X-Ray Computed , Adipose Tissue , Subcutaneous Fat , Neural Networks, ComputerABSTRACT
BACKGROUND: Estimating whole-body composition from limited region-computed tomography (CT) scans has many potential applications in clinical medicine; however, it is challenging. PURPOSE: To investigate if whole-body composition based on several tissue types (visceral adipose tissue [VAT], subcutaneous adipose tissue [SAT], intermuscular adipose tissue [IMAT], skeletal muscle [SM], and bone) can be reliably estimated from a chest CT scan only. METHODS: A cohort of 97 lung cancer subjects who underwent both chest CT scans and whole-body positron emission tomography-CT scans at our institution were collected. We used our in-house software to automatically segment and quantify VAT, SAT, IMAT, SM, and bone on the CT images. The field-of-views of the chest CT scans and the whole-body CT scans were standardized, namely, from vertebra T1 to L1 and from C1 to the bottom of the pelvis, respectively. Multivariate linear regression was used to develop the computer models for estimating the volumes of whole-body tissues from chest CT scans. Subject demographics (e.g., gender and age) and lung volume were included in the modeling analysis. Ten-fold cross-validation was used to validate the performance of the prediction models. Mean absolute difference (MAD) and R-squared (R2 ) were used as the performance metrics to assess the model performance. RESULTS: The R2 values when estimating volumes of whole-body SAT, VAT, IMAT, total fat, SM, and bone from the regular chest CT scans were 0.901, 0.929, 0.900, 0.933, 0.928, and 0.918, respectively. The corresponding MADs (percentage difference) were 1.44 ± 1.21 L (12.21% ± 11.70%), 0.63 ± 0.49 L (29.68% ± 61.99%), 0.12 ± 0.09 L (16.20% ± 18.42%), 1.65 ± 1.40 L (10.43% ± 10.79%), 0.71 ± 0.68 L (5.14% ± 4.75%), and 0.17 ± 0.15 L (4.32% ± 3.38%), respectively. CONCLUSION: Our algorithm shows promise in its ability to estimate whole-body compositions from chest CT scans. Body composition measures based on chest CT scans are more accurate than those based on vertebra third lumbar.
Subject(s)
Tomography, X-Ray Computed , Tomography , Humans , Body CompositionABSTRACT
We present a novel integrative computerized solution to automatically identify and differentiate pulmonary arteries and veins depicted on chest computed tomography (CT) without iodinated contrast agents. We first identified the central extrapulmonary arteries and veins using a convolutional neural network (CNN) model. Then, a computational differential geometry method was used to automatically identify the tubular-like structures in the lungs with high densities, which we believe are the intrapulmonary vessels. Beginning with the extrapulmonary arteries and veins, we progressively traced the intrapulmonary vessels by following their skeletons and differentiated them into arteries and veins. Instead of manually labeling the numerous arteries and veins in the lungs for machine learning, this integrative strategy limits the manual effort only to the large extrapulmonary vessels. We used a dataset consisting of 120 chest CT scans acquired on different subjects using various protocols to develop, train, and test the algorithms. Our experiments on an independent test set (nâ¯=â¯15) showed promising performance. The computer algorithm achieved a sensitivity of â¼98% in labeling the pulmonary artery and vein branches when compared with a human expert's results, demonstrating the feasibility of our computerized solution in pulmonary artery/vein labeling.
Subject(s)
Pulmonary Artery , Tomography, X-Ray Computed , Algorithms , Humans , Neural Networks, Computer , Pulmonary Artery/diagnostic imaging , Thorax , Tomography, X-Ray Computed/methodsABSTRACT
The diaphragm is an inconspicuous fibromuscular septum, and disorders may result in respiratory impairment and morbidity and mortality when untreated. Radiologists need to accurately diagnose diaphragmatic disorders, understand the surgical approaches to diaphragmatic incisions/repairs, and recognize postoperative changes and complications. Diaphragmatic defects violate the boundary between the chest and abdomen, with the risk of herniation and strangulation of abdominal contents. In our surgical practice, patients with diaphragmatic hernias present acutely with incarceration and/or strangulation. Bochdalek hernias are commonly diagnosed in asymptomatic older adults on computed tomography; however, when viscera or a large amount of fat herniates into the chest, surgical intervention is strongly advocated. Morgagni hernias are rare in adults and typically manifest acutely with bowel obstruction. Patients with traumatic diaphragm injury may have an acute, latent, or delayed presentation, and radiologists should be vigilant in inspecting the diaphragm on the initial and all subsequent thoracoabdominal imaging studies. Almost all traumatic diaphragm injury are surgically repaired. Finally, with porous diaphragm syndrome, fluid, air, and tissue from the abdomen may communicate with the pleural space through diaphragmatic fenestrations and result in a catamenial pneumothorax or large pleural effusion. When the underlying disorder cannot be effectively treated, the goal of surgical intervention is to establish the diagnosis, incite pleural adhesions, and close diaphragmatic defects. Diaphragmatic plication may be helpful in patients with eventration or acquired injuries of the phrenic nerve, as it can stabilize the affected diaphragm. Phrenic nerve pacing may improve respiratory function in select patients with high cervical cord injury or central hypoventilation syndrome.
Subject(s)
Diaphragm/injuries , Diaphragm/surgery , Hernia, Diaphragmatic/diagnostic imaging , Hernia, Diaphragmatic/surgery , Tomography, X-Ray Computed/methods , Diaphragm/diagnostic imaging , HumansABSTRACT
Esophageal surgery has become quite specialized, and both dedicated diagnostic and refined surgical techniques are required to deliver state-of-the-art care. The field has evolved to include endoscopic mucosal resection and radiofrequency ablation for early-stage esophageal cancer and minimally invasive esophagectomy with the reconstruction of a gastric conduit for carefully selected patients with esophageal cancer or those with "end-stage" esophagus from benign diseases. Reoperative esophageal surgery after esophagectomy deserves special mention given that these patients, with improved survival, are presenting years after esophagectomy with functional and anatomic disorders that sometimes require surgical intervention. Different diagnostic modalities are essential for assessing patients and planning surgical treatment. Recognizing early and late postoperative complications on imaging may expedite and improve patient outcomes. Finally, endoscopic management of achalasia with peroral endoscopic myotomy and the use of the LINX device for gastroesophageal reflux disease are highly effective and minimally invasive treatments that may reduce complications, costs, and length of hospital stay.
Subject(s)
Esophageal Diseases/surgery , Diagnostic Imaging/methods , Esophageal Diseases/diagnostic imaging , Esophagectomy/methods , Esophagus/diagnostic imaging , Esophagus/surgery , HumansABSTRACT
Herein, we present a case of successfully treated biopsy-proven Rasamsonia argillacea species complex myocarditis, pericarditis, and pulmonary infection in a 35-year-old male with a history of chronic granulomatous disease. Computed tomography of the chest demonstrated numerous pulmonary nodules and mass-like pulmonary lesions, and subsequent cardiac magnetic resonance imaging demonstrated an infiltrating mass-like lesion within the interventricular septum and pericarditis. Endobronchial, thoracoscopic, and eventual myocardial biopsies with cultures were ultimately reported as positive for R. argillacea species and the patient was treated with tailored antifungal therapy resulting in a significant therapeutic response upon short interval follow-up. This case stresses the importance of recognizing unusual thoracic imaging manifestations of an atypical fungal infection in immunocompromised individuals in order to expedite treatment of an otherwise potentially fatal disease.
ABSTRACT
There is an unexpectedly high incidence of PVOD in patients with SSc-PH-ILD. Presence of PVOD may be an unrecognised contributor to the dismal prognosis of these patients. Early transplant referral should be considered for those with SSc-PH-ILD. http://ow.ly/vPvc30neJZV.
ABSTRACT
Diagnosis of a bronchopleural fistula (BPF) can be challenging in patients after pneumonectomy and Clagett window. Herein, we present a case of pneumonectomy for advanced lung cancer complicated by a BPF. Herniation of packing material from the open-chest cavity into the fistula and airways on computed tomography was an important clue to making the diagnosis.
ABSTRACT
BACKGROUND: Catheter-directed interventions for the treatment of patients with submassive pulmonary embolism (sPE) have shown promise in rapidly improving right-sided heart strain and preventing decompensation to massive pulmonary embolism. Among various catheter interventions, ultrasound-assisted thrombolysis (USAT) has attracted interest as potentially having more efficient lytic effect that could achieve thrombolysis faster and with a reduced lytic dose. However, based on clinical evidence, it is unclear whether USAT is superior to standard catheter-directed thrombolysis (SCDT). We herein describe the study design of the Standard vs UltrasouNd-assiSted CathEter Thrombolysis for Submassive Pulmonary Embolism (SUNSET sPE) trial, an ongoing randomized clinical trial designed to address this question. METHODS: Adults with sPE presenting or referred to our institution are considered for enrollment in the trial. At the discretion of the treatment team, all patients undergo a catheter-directed intervention plus concomitant therapeutic anticoagulation. Participants are randomized 1:1 to a USAT catheter or an SCDT catheter. Study assessors are blinded to treatment group. The primary outcome is clearance of pulmonary thrombus burden, assessed by postprocedure computed tomography angiography. Secondary outcomes include resolution of right ventricular strain by echocardiography; improvement in pulmonary artery pressures; and 3- and 12-month improvement in echocardiographic, functional capacity, and quality of life measures. The study is powered to detect a 50% improvement in pulmonary artery thrombus clearance. Our enrollment target is 40 patients per treatment arm. CONCLUSIONS: SUNSET sPE is an ongoing randomized, head-to-head, single-blinded clinical trial with the goal of assessing whether USAT results in superior thrombus clearance compared with SCDT in patients with sPE. We expect the results of our study to inform future guidelines on choice of thrombolysis modality in this population of challenging patients.
Subject(s)
Fibrinolytic Agents/administration & dosage , Mechanical Thrombolysis/methods , Pulmonary Embolism/therapy , Thrombolytic Therapy , Ultrasonic Therapy , Anticoagulants/administration & dosage , Clinical Protocols , Computed Tomography Angiography , Fibrinolytic Agents/adverse effects , Humans , Mechanical Thrombolysis/adverse effects , Prospective Studies , Pulmonary Embolism/diagnostic imaging , Pulmonary Embolism/physiopathology , Quality of Life , Recovery of Function , Research Design , Single-Blind Method , Thrombolytic Therapy/adverse effects , Time Factors , Treatment Outcome , Ultrasonic Therapy/adverse effectsABSTRACT
Ulcerative colitis can cause inflammation of small and large airways, characterized by mucosal inflammation, tracheobronchial stenosis, bronchiestasis, and bronchiolitis. We present a case of tracheobronchitis and bronchiolitis associated with ulcerative colitis in a 58-year-old nonsmoking man, 17 years after the total colectomy and complete resolution of intestinal findings. Computed tomography demonstrated wall thickening of trachea and left main stem bronchus, and multiple bronchi around the both hilum with mild to moderate stenosis. Fiberoptic bronchial biopsy showed inflammation of the airways, similar to histologic findings of ulcerative colitis within colon.
ABSTRACT
Traumatic diaphragmatic rupture remains a diagnostic challenge for both radiologists and surgeons. In recent years, multidetector CT has markedly improved the diagnosis of diaphragmatic injury in polytrauma patients. Herein, we describe two cases of subacute presentation of traumatic diaphragmatic rupture from a penetrating rib fracture and subsequent intrathoracic herniation of omental fat, representing the CT "funky fat" sign.
ABSTRACT
Talcoma is a pleural mass which may develop as a rare complication following talc pleurodesis. Talc pleurodesis is performed to obliterate the pleural space to prevent recurrent pleural effusions or persistent pneumothoraces. The present report describes a case of a patient who developed enlarging pleural mass (talcoma) following talc pleurodesis.
ABSTRACT
Granulomatosis with polyangiitis is a systemic disease resulting in necrotizing vasculitis of small- and medium-sized vessels. Cardiac involvement is rare and when present usually manifests with pericarditis and coronary artery vasculitis. We report here a case of granulomatosis with polyangiitis involving the native coronary arteries, bypass graft, and pericardium with interesting imaging findings on contrast-enhanced CT and MRI. A 57-year-old man with a history of chronic headaches presented to the emergency room with syncope. Contrast-enhanced CT demonstrated extensive soft tissue attenuation around the native coronary arteries and bypass graft. Contrast-enhanced MRI demonstrated enhancing nodular soft tissue surrounding the coronary arteries, bypass graft, and pericardium. Pericardial biopsy revealed a necrotizing granulomatous pericarditis with vasculitis concerning for granulomatosis with polyangiitis. The patient demonstrated MPO-positive and PR-3 negative serologies. After being discharged on rituximab and prednisone, follow-up CT 3 years later showed significant improvement of the soft tissue thickening surrounding the coronary arteries, bypass graft, and pericardium.
ABSTRACT
OBJECTIVES: To determine if measurements of aortic wall attenuation can improve the CT diagnosis of acute aortic syndromes. METHODS: CT reports from a ten year period were searched for acute aortic syndromes (AAS). Studies with both an unenhanced and a contrast enhanced (CTA) series that had resulted in the diagnosis of intramural hematoma (IMH) were reviewed. Diagnoses were confirmed by medical records. The attenuation of aortic wall abnormalities was measured. The observed attenuation threshold was validated using studies from 39 new subjects with a variety of aortic conditions. RESULTS: The term "aortic dissection" was identified in 1206, and IMH in 124 patients' reports. IMH was confirmed in 31 patients, 21 of whom had both unenhanced and contrast enhanced images. All 21 had pathologic CTA findings, and no CTA with IMH was normal. Attenuation of the aortic wall was greater than 45 HUs on the CTA images in all patients with IMH. When this threshold was applied to the new group, sensitivity for diagnosing AAS was 100% (19/19), and specificity 94% (16/17). Addition of unenhanced images did not improve accuracy. CONCLUSIONS: Measurements of aortic wall attenuation in CTA have a high negative predictive value for the diagnosis of acute aortic syndromes.
Subject(s)
Aortic Aneurysm/diagnostic imaging , Aortic Dissection/diagnostic imaging , Aortography/methods , Atherosclerosis/diagnostic imaging , Hematoma/diagnostic imaging , Radiographic Image Interpretation, Computer-Assisted/methods , Tomography, X-Ray Computed/methods , Acute Disease , Adult , Aged , Aged, 80 and over , Female , Humans , Male , Middle Aged , Radiographic Image Interpretation, Computer-Assisted/standards , Reproducibility of Results , Sensitivity and Specificity , Syndrome , Tomography, X-Ray Computed/standards , Young AdultABSTRACT
OBJECTIVE: The purpose of this study was to assess the CT-pathologic features of cancer incidentally detected at lung transplantation. MATERIALS AND METHODS: Our lung transplant registry was reviewed over 7 years for incidental malignancy. Patient demographics, diffuse lung disease, surgical procedure, histopathology, and chest CT were recorded. We correlated lesion size, morphology, multiplicity, and location with surgical and pathology reports and histopathology. Cancers were pathologically staged. RESULTS: Of 759 lung transplant recipients, cancer was incidentally detected in 22 (2.9%). Half (11 of 258) or 4.3% were detected within the past 2 years. Four patients had a history of treated malignancy, and three had recurrence. Patients had emphysema (chronic obstructive pulmonary disease [COPD]) (n = 10), fibrosis (n = 10), or combined COPD and fibrosis (n = 2). Histopathology revealed 13 solitary lung carcinomas, four multifocal adenocarcinomas, three metastases, and two lymphoproliferative diseases. Lung cancer (n = 17) stages were I or II (n = 13), IIIA (n = 2), or IV (n = 2). Metastases (n = 3) and lymphoproliferative disease (n = 2) represented advanced disease. The interval between CT and surgery was a mean of 4 months. CT-positive cases (n = 10) represented lung cancer (n = 9) and posttrans-plantation lymphoproliferative disease (n = 1). Cases with no CT findings of malignancy (n = 12) included lung cancer (n = 8), metastases (n = 3), and lymphoma (n = 1). Ten cases (45%) had other histologically benign CT abnormalities that mimicked cancer. CONCLUSION: Detection of incidental malignancy at lung transplantation has increased over the past 2 years. Malignancies were typically stage I or II lung cancers that were occult or indeterminate on CT. Diffuse lung disease, multiple CT abnormalities, and a delay between CT and transplantation compromise the preoperative diagnosis of cancer.
Subject(s)
Lung Neoplasms/diagnostic imaging , Lung Transplantation , Lymphoproliferative Disorders/diagnostic imaging , Tomography, X-Ray Computed/methods , Adult , Aged , Female , Humans , Incidental Findings , Lung Neoplasms/pathology , Lymphoproliferative Disorders/pathology , Male , Middle Aged , Neoplasm Staging , Preoperative Period , RegistriesABSTRACT
Lipomatosis is a benign and non-encapsulated mature form of fat tissue proliferation, which can be isolated or multiple, symmetrical or infiltrative, and may be associated with several syndromes. Lipomatosis has been reported to involve almost every part and organ in the body. In this review, we present imaging findings at common locations of lipomatosis and summarize lipomatosis-related syndromes.
Subject(s)
Diagnostic Imaging , Lipomatosis/diagnosis , Diagnosis, Differential , Humans , SyndromeABSTRACT
BACKGROUND: Idiopathic pulmonary fibrosis (IPF) is a progressive and medically refractory lung disease with a grim prognosis. Although the etiology of IPF remains perplexing, abnormal adaptive immune responses are evident in many afflicted patients. We hypothesized that perturbations of human leukocyte antigen (HLA) allele frequencies, which are often seen among patients with immunologic diseases, may also be present in IPF patients. METHODS/PRINCIPAL FINDINGS: HLA alleles were determined in subpopulations of IPF and normal subjects using molecular typing methods. HLA-DRB1*15 was over-represented in a discovery cohort of 79 Caucasian IPF subjects who had lung transplantations at the University of Pittsburgh (36.7%) compared to normal reference populations. These findings were prospectively replicated in a validation cohort of 196 additional IPF subjects from four other U.S. medical centers that included both ambulatory patients and lung transplantation recipients. High-resolution typing was used to further define specific HLA-DRB1*15 alleles. DRB1*1501 prevalence in IPF subjects was similar among the 143 ambulatory patients and 132 transplant recipients (31.5% and 34.8%, respectively, pâ=â0.55). The aggregate prevalence of DRB1*1501 in IPF patients was significantly greater than among 285 healthy controls (33.1% vs. 20.0%, respectively, OR 2.0; 95%CI 1.3-2.9, pâ=â0.0004). IPF patients with DRB1*1501 (nâ=â91) tended to have decreased diffusing capacities for carbon monoxide (DL(CO)) compared to the 184 disease subjects who lacked this allele (37.8±1.7% vs. 42.8±1.4%, pâ=â0.036). CONCLUSIONS/SIGNIFICANCE: DRB1*1501 is more prevalent among IPF patients than normal subjects, and may be associated with greater impairment of gas exchange. These data are novel evidence that immunogenetic processes can play a role in the susceptibility to and/or manifestations of IPF. Findings here of a disease association at the HLA-DR locus have broad pathogenic implications, illustrate a specific chromosomal area for incremental, targeted genomic study, and may identify a distinct clinical phenotype among patients with this enigmatic, morbid lung disease.